In R, how to make the jitter (geom_jitter()) stay inside its correspondant boxplot without extending over the neighboring boxplots?

Question

I would like to find a way for the jitter to stay in its own boxplot, without extending over the neighboring boxplots.

So far, I looked at this answers:

• R- Group jitter in factored boxplot?
• Understanding boxplot with 'jitter'
• ggplot2 - jitter and position dodge together

but none of them really addressed my issue; the main difference is that I have 3 groups running through a timeline on the X-axis.

The code I have so far:

ggplot(longitudinal, mapping= aes(x = Time, y = Values), shape= Diagnose)+
geom_boxplot(aes(color = Diagnose), outlier.shape = NA ) +
geom_jitter(aes(color= Diagnose, shape=Diagnose)  ,alpha = 0.5)

The image output:

As you can see, the jitter obeys the Timepoint distribution (T0, T1, T2, T3), but when it comes to the diagnosis(Diagnose), it overlaps with the other boxes.

Here is an example of how my data looks like:

structure(list(Time = c("T0", "T0", "T0", "T0", "T0", "T0", "T0", 
"T0", "T0", "T1", "T1", "T1", "T1", "T1", "T1", "T1", "T1", "T2", 
"T2", "T2", "T2", "T2", "T2", "T2", "T2", "T2", "T3", "T3", "T3", 
"T3", "T3", "T3", "T3", "T3", "T3"), Diagnose = c("PDD", "PDD", 
"PDD", "PD-MCI", "PD-MCI", "PD-MCI", "PD", "PD", "PD", "PD", 
"PD", "PD-MCI", "PD-MCI", "PD-MCI", "PDD", "PDD", "PDD", "PD", 
"PD", "PD", "PD-MCI", "PD-MCI", "PD-MCI", "PDD", "PDD", "PDD", 
"PD", "PD", "PD", "PD-MCI", "PD-MCI", "PD-MCI", "PDD", "PDD", 
"PDD"), Values = c(13.47, 14.25, 15, 20, 19.57, 15, 15, 17.54, 
18, 16.93, 11.42, 18, 15, 19.48, 15, 11, 15, 18.03, 11, 15, 17.85, 
19, 15, 15, 17.85, 20, 15, 19, 14.11, 12, 18.31, 16, 17.36, 20, 
12)), row.names = c(NA, -35L), class = c("tbl_df", "tbl", "data.frame"
))

and this the output when using position = position_jitter(), position=position_jitterdodge(), position_dodge, position_jitterdodge(dodge.width= ) etc... As you can see, this packs all the jitter in the central boxplots.

Thanks!

Answer 1

Almost! what you are looking for is geom_point(position = position_jitterdodge()) . You can also adjust the width with jitter.width

 ggplot(df, mapping= aes(x = Time, y = Values))+
  geom_boxplot(aes(color = Diagnose), outlier.shape = NA ) +
  geom_point(aes(color= Diagnose, shape=Diagnose), alpha = 0.5, 
                 position = position_jitterdodge(jitter.width = 0.1))

Answer 2

Specify the dodge width

+ geom_jitter(width = 0.05)

or geom_point(position = position_jitter(width = 0.05))